In recent years, investigations have been actively advanced for applying an ionic liquid as an electrolytic solution for batteries or electric double layer capacitors, and for using an ionic liquid as a plating bath. In conventional batteries or electric double layer capacitors, an aqueous electrolytic solution or an organic electrolytic solution has been used as an electrolytic solution. However, the aqueous electrolytic solution has the problem of being restricted about the decomposition voltage of water. The organic electrolytic solution has problems about heat resistance and safety. By contrast, an ionic liquid has characteristics preferable for safety, such as flame resistance and nonvolatility, and is also high in electrochemical stability. The ionic liquid is therefore suitable, in particular, for an electrolytic solution for batteries or electric double layer capacitors used in a high-temperature environment.
In order to use an ionic liquid as an electrolytic solution for batteries or electric double layer capacitors, investigations have been advanced about various types of ionic liquids each composed of a cation and an anion. For example, Non-Patent Document 1 reports properties of 1-ethyl-3-methylimidazolium difluorophosphate, which has a difluorophosphate as an anion, as an ionic liquid. Non-Patent Document 2 reports that the 1-ethyl-3-methylimidazolium difluorophosphate has electroconductivity and voltage resistance equivalent to those of 1-ethyl-3-methylimidazolium tetrafluoroborate, which is known as a typical ionic liquid, and can be suitably used for an electrolyte for electric double layer capacitors.
According to Non-Patent Document 1, in a method for producing the 1-ethyl-3-methylimidazolium difluorophosphate, the 1-ethyl-3-methylimidazolium difluorophosphate can be produced by reacting 1-ethyl-3-methylimidazolium chloride with potassium difluorophosphate in acetone, filtering off potassium chloride produced as a by-product from the solution in acetone, allowing the remaining solution to act onto an alumina column, and then distilling away acetone therefrom. Impurities in an electrolytic solution remarkably affect performances of batteries or electric double layer capacitors; thus, when an ionic liquid is used as an electrolytic solution, it is preferred to reduce impurities as much as possible. The ionic liquid is hardly volatile, and is also in a liquid state within a broad temperature range, so that the impurities are not easily reduced by a purifying method such as distillation or recrystallization. It is therefore necessary for synthesizing a high-purity ionic liquid to use a high-purity raw material. Thus, it is desired in the production method disclosed in Non-Patent Document 1 that the amount of impurities contained in potassium difluorophosphate to be used is as small as possible.
Methods for producing a difluorophosphate such as potassium difluorophosphate are disclosed in, for example, Patent Documents 1 to 8 and Non-Patent Documents 3 to 7 listed below.
Non-Patent Documents 3 and 4 each disclose a method of allowing ammonium fluoride or acidic sodium fluoride to act onto diphosphorous pentaoxide to provide a difluorophosphate. However, in the respective production methods disclosed in these documents, besides the difluorophosphate, a monofluorophosphate, a phosphate, and water are produced as by-products in large amounts. Accordingly, a large burden is imposed on a subsequent purifying step. Thus, it is not easily mentioned that these methods are effective methods.
Non-Patent Document 5 discloses a method of allowing P2O3F4 (difluorophosphoric anhydride) to act onto, for example, an oxide or hydroxide such as Li2O or LiOH to produce a difluorophosphate. However, difluorophosphoric anhydride is very expensive, and high-purity difluorophosphoric anhydride is not easily available. Thus, this production method is disadvantageous for industrial production.
Patent Document 1 discloses a method of mixing potassium hexafluorophosphate with potassium metaphosphate, and melting the mixture to provide potassium difluorophosphate. However, this production method has the following problem: potassium difluorophosphate is contaminated by a crucible used at the time of melting potassium hexafluorophosphate and potassium metaphosphate. For the melting, it is also necessary to realize an environment of a high temperature such as 700° C. From the viewpoints of product purity and productivity, the production method disclosed in Patent Document 1 cannot be said to be a preferable method.
Non-Patent Document 6 discloses a method of melting urea, potassium dihydrogenphosphate, and ammonium fluoride to react these compounds with one another, thereby producing potassium difluorophosphate. In this production method, the reaction temperature can be lowered to about 170° C. In light of a comparison of this condition with reaction conditions in Patent Document 1, this production method makes it possible to realize industrial production. However, there remain the following problems: it is necessary to dispose of a large amount of ammonia gas produced as a by-product, and a large amount of ammonium fluoride also remains. Thus, from the viewpoints of production efficiency and product purity, the production method disclosed in Non-Patent Document 6 is not preferable, either.
Non-Patent Document 7 discloses a method of: reacting an alkali metal chloride with excessive difluorophosphoric acid; heating and drying hydrogen chloride, which is produced as a by-product, and a surplus of difluorophosphoric acid under reduced pressure to be distilled away; and then obtaining a difluorophosphate. However, even when difluorophosphoric acid sufficiently high in purity is used, a monofluorophosphate and a fluoride salt remain as impurities in large amounts in the difluorophosphate obtained by this method. It is therefore also difficult that the production method disclosed in Non-Patent Document 7 provides a high-purity difluorophosphate.
Patent Documents 2 to 4 each disclose a method of reacting lithium hexafluorophosphate with a borate, silicon dioxide and a carbonate in a nonaqueous solvent to provide lithium difluorophosphate. Moreover, Patent Document 5 discloses a method of bringing a carbonate or borate into contact with a gas such as phosphorous pentafluoride to provide lithium difluorophosphate. However, the production methods disclosed in these documents require a process over a long time of, for example, 40 to 170 hours for providing a difluorophosphate. Thus, these methods are unsuitable for industrial production.
Patent Document 6 describes a method of reacting an oxoacid or oxyhalide of phosphorous with a hexafluorophosphate, a halide of an alkali metal, and the like in the presence of hydrogen fluoride to provide a difluorophosphate. According to this method, the hexafluorophosphate acts, through the presence thereof, effectively onto contaminated water so that a high-purity difluorophosphate can be obtained. However, the hexafluorophosphate, which is expensive, is used in a relatively large amount, and further according to methods described in Examples therein, an exhaust gas or waste fluid containing a large amount of phosphorous and fluorine is generated to cause the following problem: the separation and recovery of useful substances, and waste disposal are complicated.
Patent Document 7 discloses a method of reacting a halide of an alkali metal or the like with difluorophosphoric acid in the presence of a hexafluorophosphate to produce a difluorophosphate. Patent Document 8 discloses a method of reacting difluorophosphoric acid with a halide or the like of an alkali metal in difluorophosphoric acid, and providing a difluorophosphate in difluorophosphoric acid by a crystallizing operation. In these production methods, it is necessary to use a high-purity difluorophosphoric acid. However, difluorophosphoric acid is high in corrosive property; thus, reduced pressure distillation or the like is required, and further facilities for the production are complicated. There is also caused the following problem: it is difficult to industrially gain difluorophosphoric acid regardless of the purity thereof.
In the meantime, a high-purity difluorophosphate can be used not only as a raw material of an ionic liquid but also as an additive for an electrolytic solution for lithium secondary batteries. In recent years, as a field to which lithium secondary batteries are applied enlarges from that of electronic instruments such as portable phones, personal computers and digital cameras to that of articles mounted on automobiles, a further rise in performances thereof has been advanced, for example, the power density and the energy density are improved, and a loss in the capacity is restrained. Lithium secondary batteries used, particularly, in articles mounted in automobiles may be exposed to a severer environment than ones used in consumer products; accordingly, the batteries are required to have a high reliability in terms of cycle life and storage performance. An electrolytic solution used in lithium secondary batteries is a nonaqueous electrolytic solution in which a lithium salt is dissolved in an organic solvent. The decomposition of such a nonaqueous electrolytic solution, and a side reaction thereof affect the performance of the lithium secondary batteries. Consequently, attempts have been made for improving the batteries in cycle life and storage performance by mixing various additives with the nonaqueous electrolytic solution.
For example, Patent Document 9 discloses that an organic solvent, as a nonaqueous electrolytic solution for lithium secondary batteries, contains at least one of lithium monofluorophosphate and lithium difluorophosphate as an additive. Patent Document 9 states that the use of such a nonaqueous electrolytic solution makes it possible to form a film onto a positive electrode and a negative electrode, respectively, thereby restraining the electrolytic solution from being decomposed by contact between the nonaqueous electrolytic solution, and a positive active material and a negative active material. Thus, the battery can be restrained from undergoing self-discharge and be improved in storage performance.